Google uses the same search algorithm most language, but in some cases a language requires a different algorithm to interpret the query.
Here’s a response to a Reddit thread by Google’s search advocate John Mueller titled: “Does Google use the same algorithm for every language?”
The thread goes on to ask about ranking factors and SEO practices, and how they go from one language to another.
More specifically, the thread reads:
“BERT updates are about semantics, so I wondered if it was the same in every language. This got me thinking about other ranking factors and how their importance might vary across languages/cultures. Whatever How, I want to ask anyone with SEO experience in another language, do you see any difference between ranking factors?”
Mueller doesn’t address the ranking factor aspect, but does address the use of search algorithms in different languages.
Read his full response in the section below.
How Google’s search algorithm varies by language
While many refer to the Google search algorithm as a single entity, it is actually made up of “many, many” algorithms.
Some of these algorithms are used for searches in all languages, while some are used only in individual languages.
For example, Mueller said, some languages don’t separate words with spaces. This makes it necessary to use a different algorithm than the language Google uses to use spaces.
He said:
“Mostly. Searches use many, many algorithms. Some of them apply to content in all languages, some of them are specific to individual languages (eg some languages don’t use spaces to separate words – if Google assumes all languages are like English).”
source: Reddit
How Google Search understands content in different languages
On the topic of searching Google in different languages, it’s worth noting a point raised during the Google Search Central SEO office hour meetup last week.
Mueller was asked how Google can determine when one page is similar to another when each page has content in a different language.
In short, Google can not Detect when content written in one language is the same or similar to content written in another language.
Google relies on content publishers to recognize that multiple pieces of content written in different languages are equivalent.
This is achieved through the hreflang HTML attribute, Mueller explained:
“…we basically use hreflang to know which of these URLs are equivalent from your perspective. We’re going to swap them out…
…I don’t think it’s possible for us to understand that this particular content is equivalent to another country or another language. Like, there are always a lot of local differences that are possible. “
Knowing that Google cannot determine the equivalence of content in different languages on its own, it makes it clearer why Google has certain algorithms for certain languages.
source: Google Search Center on YouTube
Featured image: YummyBuum/Shutterstock
